International variation in the usage of medicines - RAND Corporation

1 downloads 204 Views 371KB Size Report
service of the RAND Corporation. ▾ ... The RAND Corporation is a nonprofit research ... the Steering Group “Extent a
THE ARTS CHILD POLICY CIVIL JUSTICE

This PDF document was made available from www.rand.org as a public service of the RAND Corporation.

EDUCATION ENERGY AND ENVIRONMENT

Jump down to document6

HEALTH AND HEALTH CARE INTERNATIONAL AFFAIRS NATIONAL SECURITY POPULATION AND AGING PUBLIC SAFETY SCIENCE AND TECHNOLOGY SUBSTANCE ABUSE

The RAND Corporation is a nonprofit research organization providing objective analysis and effective solutions that address the challenges facing the public and private sectors around the world.

TERRORISM AND HOMELAND SECURITY TRANSPORTATION AND INFRASTRUCTURE WORKFORCE AND WORKPLACE

Support RAND Browse Books & Publications Make a charitable contribution

For More Information Visit RAND at www.rand.org Explore RAND Europe View document details

Limited Electronic Distribution Rights This document and trademark(s) contained herein are protected by law as indicated in a notice appearing later in this work. This electronic representation of RAND intellectual property is provided for non-commercial use only. Unauthorized posting of RAND PDFs to a non-RAND Web site is prohibited. RAND PDFs are protected under copyright law. Permission is required from RAND to reproduce, or reuse in another form, any of our research documents for commercial use. For information on reprint and linking permissions, please see RAND Permissions.

This product is part of the RAND Corporation technical report series. Reports may include research findings on a specific topic that is limited in scope; present discussions of the methodology employed in research; provide literature reviews, survey instruments, modeling exercises, guidelines for practitioners and research professionals, and supporting documentation; or deliver preliminary findings. All RAND reports undergo rigorous peer review to ensure that they meet high standards for research quality and objectivity.

International variation in the usage of medicines A review of the literature Ellen Nolte, Jennifer Newbould, Annalijn Conklin Prepared for the Department of Health within the PRP project "An 'On-call' facility for International Healthcare Comparisons"

EUROPE

The research described in this report was prepared for the Department of Health within the PRP project "An 'On-call' Facility for International Healthcare Comparisons" (grant no. 0510002).

The RAND Corporation is a nonprofit research organization providing objective analysis and effective solutions that address the challenges facing the public and private sectors around the world. RAND’s publications do not necessarily reflect the opinions of its research clients and sponsors.

R® is a registered trademark.

© Copyright 2010 by RAND and London School of Hygiene and Tropical Medicine

All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the copyright holders.

Published 2010 by the RAND Corporation 1776 Main Street, P.O. Box 2138, Santa Monica, CA 90407-2138 1200 South Hayes Street, Arlington, VA 22202-5050 4570 Fifth Avenue, Suite 600, Pittsburgh, PA 15213-2665 Westbrook Centre, Milton Road, Cambridge CB4 1YG, United Kingdom RAND URL: http://www.rand.org RAND Europe URL: http://www.rand.org/randeurope To order RAND documents or to obtain additional information, contact Distribution Services: Telephone: (310) 451-7002; Fax: (310) 451-6915; Email: [email protected]

Preface

This report reviews the published and grey literature on international variation in the use of medicines, focusing on osteoporosis, atypical anti-psychotics, dementia, rheumatoid arthritis, cardiovascular disease/lipid-regulating drugs (statins), and hepatitis C. The report aims to inform the Steering Group “Extent and Causes of International Variation in Drug Usage” to guide further analytical work on the extent and causes of international variation in drug usage. The report was prepared as part of the project “An ‘On-call’ Facility for International Healthcare Comparisons” funded by the Department of Health in England through its Policy Research Programme (grant no. 0510002). The project comprises a programme of work on international health care comparisons that provides intelligence on new developments in other countries, involving a network of experts in a range of OECD countries to inform health policy development in England. It is conducted by RAND Europe, in conjunction with the London School of Hygiene & Tropical Medicine. RAND Europe is an independent not-for-profit research organisation whose mission is to improve policy and decision making through research and analysis. RAND Europe’s clients include European governments, institutions, NGOs, and firms with a need for rigorous, independent, multidisciplinary analysis. The London School of Hygiene & Tropical Medicine is Britain’s national school of public health and a leading postgraduate institute worldwide for research and postgraduate education in global health. For more information about this report, please contact: Dr Ellen Nolte Director, Health and Healthcare RAND Europe Westbrook Centre, Milton Road Cambridge CB4 1YG Email: [email protected] Tel: +44 (0)1223 273853

iii

Contents

Preface ................................................................................................................................... iii Tables ................................................................................................................................... vii Executive summary .................................................................................................................ix Acknowledgements .................................................................................................................xi CHAPTER 1

Background................................................................................................. 1

CHAPTER 2

Osteoporosis ............................................................................................... 3

CHAPTER 3

Atypical anti-psychotics .............................................................................. 5

CHAPTER 4

Dementia .................................................................................................. 11

CHAPTER 5

Rheumatoid arthritis ................................................................................. 15

CHAPTER 6

Cardiovascular disease lipid-regulating drugs/statins ................................ 20

CHAPTER 7

Hepatitis C ................................................................................................ 27

CHAPTER 8 Summary and conclusions ......................................................................... 29 8.1 Macro- or system-level determinants ........................................................................... 30 8.2 Service organisation determinants ............................................................................... 31 8.3 Clinical practice determinants..................................................................................... 31 REFERENCES .................................................................................................................. 33 Reference List ....................................................................................................................... 34 APPENDICES .................................................................................................................. 37 Appendix A ........................................................................................................................... 39 Appendix B ........................................................................................................................... 41

v

Tables

Table 1 Summary of studies of international variation in the use of atypical antipsychotics (AAPs) .................................................................................................... 8 Table 2 Summary of studies of international variation in the use of medicines for the treatment of dementia ........................................................................................... 12 Table 3 Summary of studies of international variation in the use of medicines for the treatment of rheumatoid arthritis (RA) .................................................................. 17 Table 4 Use of statins in European countries in 2000............................................................... 21 Table 5 Summary of studies of international variation in the use of statins ............................... 25 Table 6 Summary of studies of international variation in the use of new anti-viral drugs for the treatment of hepatitis C .................................................................... 28

vii

Executive summary

In July 2009 the Department of Health established a Steering Group “Extent and Causes of International Variation in Drug Usage” to guide analytical work to better understand the extent and causes of international variation in drug usage. This report aims to contribute to this process by reviewing the published and grey literature on international variation in the use of medicines in six areas (osteoporosis, atypical anti-psychotics, dementia, rheumatoid arthritis, cardiovascular disease/lipid-regulating drugs (statins), and hepatitis C). The systematic search found surprisingly few international comparative studies that examined medicines use and these varied widely in terms of quality and focus, populations and time periods studied, and outcomes measured. However, despite this variation several common issues emerged from the evidence reviewed here. We identify three broad groups of determinants of international variation in medicines use: •

Macro- or system level factors. Differences in reimbursement policies, and the role of health technology assessment, were highlighted as a likely driving force of international variation in almost all areas of medicines use reviewed here, including dementia, rheumatoid arthritis, hepatitis C, and, for some countries in central and eastern Europe, statins. A related but rarely studied aspect is patient co-payment, potentially explaining some of the international variation in medicines use, which is likely to play an important role in the United States in particular, compared with European countries; but the extent to which cost-sharing policies impact on overall use of medicines in international comparison remains unclear.



Service organisation and delivery. Most studies reviewed here pointed to differences in access to specialists as a likely driver of international variation in areas such as atypical antipsychotics, dementia, and rheumatic arthritis, with for example access to and availability of relevant specialists identified as acting as a crucial bottleneck for accessing treatment for dementia and rheumatoid arthritis.



Clinical practice. Several studies highlighted the role of variation in the use and ascertainment methods for mental disorders, and differences in the use of clinical or practice guidelines. Many studies further pointed to differences in prescribing patterns as an important factor, along with a potential reluctance among clinicians in some countries to take up newer medicines, but none of the studies presented here provided empirical evidence to support this notion.

Each of these factors is likely to play a role in explaining international variation in medicines use, but their relative importance will vary depending on the disease area in question and the system context. It is likely that any given level of use of a given medicine in one country is determined by

ix

International variation in the usage of medicines

International Healthcare Comparisons

a set of factors the combination and the relative weight of which will be different in another country.

x

Acknowledgements

The project “An ‘On-call’ Facility for International Healthcare Comparisons” is funded by the Department of Health in England through its Policy Research Programme (grant no. 0510002). The authors gratefully acknowledge the valuable comments provided by Martin Roland and Jonathan Grant on an earlier draft of this report. The views expressed in this report brief are those of the authors alone and do not necessarily represent those of the Department of Health. The authors are fully responsible for any errors.

xi

CHAPTER 1

Background

The 2008 review Improving Access to Medicines for NHS Patients highlighted that there is a perception among stakeholders that usage of new medicines is low in England when compared with other countries, especially for new anti-cancer drugs.1 However, building on preliminary analyses of international data on usage of medicines for the treatment of selected disease areas including cancer, the review pointed to the considerable (technical) challenges related to comparing international medicines usage and that the extent, causes and implications of such variations are not fully understood. One of the recommendations set out by the review therefore highlighted the need to examine international variation in medicines usage more systematically so as to inform future policy decisions on funding for drugs.1 Subsequent to the review the Department of Health established a Steering Group “Extent and Causes of International Variation in Drug Usage” to guide further analytical work on the extent and causes of international variation in drug usage, which began in the context of the 2008 review. This report aims to contribute to this process by reviewing the published and grey literature on international variation in the use of medicines in selected areas. This report focuses on six (disease) areas: (1) osteoporosis, (2) atypical anti-psychotics, (3) dementia, (4) rheumatoid arthritis, (5) cardiovascular disease/lipid-regulating drugs (statins), and (6) hepatitis C. The review is based on a systematic search of PubMed, complemented by a targeted search of websites of (inter)national organisations considered relevant to the medicine and/or condition under review, including the European Medicines Agency (EMEA) and the European Federation of Pharmaceutical Industries and Associations (EFPIA). The review concentrates on evidence of international variation in medicines usage and we therefore only consider studies that explore cross-national variation involving at least two countries. In order to capture the potentially varied literature we applied broad search terms, using combinations of (“/” indicating “or”) “use/utilization/consumption/ prevalence”, “therapy/treatment/medication”, “international/cross-national/Europe/ America”, and “variation/difference/comparison” linked to the individual medicine or medicine class (e.g. “alendronate” to represent bisphosphonates for the treatment of osteoporosis; “atypical antipsychotics”; “statins”) for the treatment of the disease/conditions and/or to the actual disease (e.g. “rheumatoid arthritis”). The search was restricted to selected new and/or high-cost medicines in each of the seven areas as identified by the Steering Group and listed in Appendix A. Appendix B provides a description of the search terms and strategies used for each area. The search was limited to studies published from 2000 onwards. We imposed this restriction mainly because the majority of medicines reviewed here were only licensed in the 2000s. Furthermore, pharmaceutical policies that are likely to exert an influence on drug usage have

1

International variation in the usage of medicines

International Healthcare Comparisons

changed considerably in a number of countries during the past decade, with for example the introduction of reference pricing, parallel imports, or the use of negative/positive lists; an example is the impact of the introduction of reference pricing on the use of statins in Germany.2 3 Studies that predate these policies are unlikely to inform a better understanding of causes of contemporary variation in medicines usage. Titles and abstracts were screened for eligibility for inclusion. Studies considered eligible were retrieved where possible and scrutinised further for inclusion or exclusion in the review. References in studies considered eligible were followed up where appropriate. We generally excluded studies that examined the (cost-) effectiveness and/or safety of the medicines in question, clinical and drug trials, etiological (observational) studies, and those that examined patient adherence. We also excluded review articles except as a source for the identification of international comparative studies not identify by our search. Overall the systematic search identified surprisingly few studies of international variation in the use of the medicines under review and none for osteoporosis. In the latter case we therefore also considered studies of patient preferences that, while not directly comparing usage of the listed medicines for the treatment of osteoporosis (Appendix A), provided some indirect evidence that could offer insights into the possible causes for international variation in the usage of these medicines. We included only studies comparing medicines usage in high-income/OECD countries, with a particular focus on Europe, Canada, and the United States. We thus excluded studies comparing medicines usage among the devolved countries of the United Kingdom. Studies in languages other than English, German, or French were also excluded. Eligible studies were further analysed by extracting data according to a common template: •

stated study objective



study design (type of study)



year



population/s studied



data source



outcome measure/s



key findings.

Where appropriate, we included a brief commentary to highlight specifics of the study in question that are likely to influence the generalisability of the findings (e.g. data are not population based).

2

CHAPTER 2

Osteoporosis

The PubMed search identified 317 records of which none was considered eligible for inclusion according to the inclusion criteria set out in the preceding section. We further reviewed the websites of the European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis, the International Osteoporosis Foundation, and the European Federation of Pharmaceutical Industries and Associations (EFPIA). This search yielded two related reports on the management of osteoporosis in the 27 Member States of the European Union in 2008.4 5 The report focused, among other things, on access to diagnostics and treatment in each Member State, but did not analyse uptake or usage of medicines for the treatment of osteoporosis and so was not included in our review. An earlier scanning of the literature undertaken to inform the search strategy used here identified one study that examined women’s treatment preferences for osteoporosis in different countries.6 Duarte et al. (2007) undertook a cross-sectional survey of postmenopausal women with osteoporosis in France, Germany, Mexico, Spain and the United Kingdom (n=600 in each country) to determine the relative importance of seven medication attributes in assessing participant preferences for osteoporosis medication. The importance women placed on different attributes differed among countries, with effectiveness rated as most important in their preference for a prescription osteoporosis medication ranging from 49 per cent in Mexico to 71 per cent in France. The role of out-of-pocket payments was generally rated as of low importance in France, Germany, Spain and the United Kingdom (at under 8 per cent of all women surveyed), but not Mexico (16 per cent). Among women treated for osteoporosis, alendronate was the most commonly used medication in most countries (26 per cent of all respondents), from 14 per cent in Mexico to 34 per cent in Germany.6

3

CHAPTER 3

Atypical anti-psychotics

The PubMed search identified 205 records of which seven studies met our inclusion criteria. We further searched the websites of Mind; the Prescribing Observatory for Mental Health, hosted by the Royal College of Psychiatrists (UK); the Mental Health Foundation; the Working Group Drugs in Psychiatry (AGATE); and the European Federation of Pharmaceutical Industries and Associations (EFPIA). This search did not identify any additional comparative studies that met our inclusion criteria. Of the seven studies that included two or more countries with regard to the use of atypical antipsychotics (AAPs), one pooled pharmacists’ data on AAP use and polypharmacy in 45 hospitals in six countries (Belgium, Denmark, France, Germany, the Netherlands and Scotland).7 It found that frequencies of atypical polypharmacy ranged between 26.1 per cent (Germany) and 49.1 per cent (Belgium) (Table 1). However, the authors noted that the study was not designed to analyse variation in the use of atypical anti-psychotics, with several countries underrepresented in their study and therefore findings have to be interpreted with caution. Santamaria et al. (2002) analysed trends in the use of anti-psychotics in Spain between 1985 and 2000 using sales data, and then compared these to published sales data in five Nordic countries (Table 1).8 However, the comparison with the Nordic countries does not allow for conclusions about cross-national trends for consumption of AAPs specifically. Zito et al. (2008) compared the annual prevalence in 2000 of any psychotropic medication in young people (0–19 years) in the Netherlands, Germany and the United States (Table 1).9 They found relatively modest cross-national differences in the prevalence of total anti-psychotics (from 0.34 per 100 youth in Germany to 0.76 per 100 youth in the United States), but the proportion of AAPs in relation to total anti-psychotic use was much lower in Germany (5 per cent) than in the two comparator countries (48 per cent in the Netherlands and 66 per cent in the United States). The authors discussed a number of factors that might explain observed variations in psychotropic medication in young people in the three countries although they did not distinguish factors specifically related to AAP use. Potential explanatory factors put forward include differences in diagnostic classification between the United States (using Diagnostic and Statistical Manual (DSM) criteria) and European countries (application of the International Classification of Diseases version 10); higher prevalence of concomitant use of two or more medications in the United States; and access to specialists such as child psychiatrists, which tended to be lower in European countries; however, the authors did not provide empirical evidence in support of this discussion.9 Four studies analysed cross-national variation in AAP use among selected, psychiatric (in)patient populations in different countries, usually focusing on prescription patterns. Thus, Hübner-

5

International variation in the usage of medicines

International Healthcare Comparisons

Liebermann et al. (2005) compared inpatient treatment of patients with schizophrenia in two hospitals in Germany and Japan, using data from the psychiatric basic documentation system in 1997.10 The rate of usage of AAPs was 18 per cent at the German hospital and 12 per cent at the Japanese hospital (Table 1). Potential explanations for the observed variation put forward by the authors include differences in service provision and national psychiatric practices. The authors also commented on availability and affordability as a possible cause for country variation since AAPs in 1997 were rare in Japan as elsewhere in the Asian region, except for zotepine. However, as their study only compared single hospitals in each country, the findings are not easily generalisable. Zullino et al. (2005) analysed psychotropic drug prescriptions, including of AAPs, in five German and Swiss psychiatric hospitals (with a total of 572 patients) in 1999.11 Although the proportion of patients receiving antidepressants varied significantly between the two countries, with a higher propensity among Swiss clinicians than clinicians at the three German hospitals to prescribe psychotropic drugs (at 65.2 per cent compared with 48.3 per cent), AAP prescriptions did not vary between Germany and Switzerland. However, a higher proportion of patients in Germany received clozapine (at 70.8 per cent of prescribed AAP) than in Switzerland (43.9 per cent). In discussing their findings, the authors largely focused on potential explanations for the observed variation in antidepressants, including differences in the patient population, hospital-specific factors, and cost considerations. There was little discussion of the observed lack of such a variation in AAP prescribing although the higher proportion of prescribing clozapine in German hospitals might point to possible lack of uptake of newer medicines in Germany. However, it is important to stress that although the five hospitals included in the study were selected to represent university and non-university settings, as well as urban and rural areas, generalisability of findings is limited. Haro et al. (2003) analysed the baseline characteristics of patients recruited into the Schizophrenia Outpatient Health Outcomes (SOHO) study in ten western European countries in 2000 and 2001 (Denmark, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain and United Kingdom).12 Atypical anti-psychotic treatment in the six months prior to recruitment to the study varied between countries, from a high of 60.6 per cent in Denmark and 58.1 per cent in Portugal to a low of 34.2 per cent in Germany and 27.4 per cent in Greece. There was considerable variation in the use of specific AAPs at baseline, with usage rates (defined as receiving an AAP) of risperidone of 9.1 per cent in Ireland; around 13 per cent in Germany and Portugal; 16 per cent in the United Kingdom; 18 per cent Denmark and Italy; 22 per cent in France, the Netherlands and Spain; and 30.7 per cent in Greece. Use of amisulpride varied from 16 per cent in Ireland; to 10 per cent in France; to no use in Denmark, the Netherlands, Greece or Spain, where the drug was not available at that time (2001/02). The use of clozapine ranged from a high of 12.1 per cent in Denmark; to between 5 per cent and 6 per cent in Ireland, the United Kingdom and Italy; 3.6 per cent in the Netherlands; 2.7 per cent in Germany; 1–2 per cent in Portugal, Spain and Greece; and 0.5 per cent in France. The authors noted that these variations are not likely to reflect differences in sociodemographic or clinical patient characteristics, but represent treatment differences across countries, listing local prescribing practices, availability of treatment guidelines, and cost as potential variables that

6

International Healthcare Comparisons

Atypical anti-psychotics

might explain observed differences in patterns of use. However, their study was not designed to explore variation in treatment patterns. Finally, Bitter et al. (2003) reported on the findings of a multi-centre comparative study of prescribing practices for acute inpatients with schizophrenia (429 patients) in six academic psychiatric departments in China, Japan, Hungary, and the United States in 1999.13 There was considerable variation between centres regarding the proportion of patients receiving AAPs, ranging from a high of 73–75 per cent in one Hungarian centre (Budapest 1) and China (Shanghai) to a low of just under 9 per cent in the Japanese centre (Tokyo) (Table 1). However, these differences were not statistically significant when adjusted for differences in demographic variables. The authors highlighted several limitations of their study, including representativeness of the patients, prescribers and hospitals, so findings are not easily generalisable. Although these four studies provide important insights into possible international variation in usage of AAPs, the representativeness of findings is somewhat limited given the design of each study, involving only single or a low number of medical units in the relevant countries under review and/or the study was not set up to explore variation in AAP use.

7

International variation in the usage of medicines

International Healthcare Comparisons

Table 1 Summary of studies of international variation in the use of atypical anti-psychotics (AAPs) Stated study objective

Study design

Year

Data source

Population/s studied

Outcome measure/s

Key findings

Notes

Source

To investigate the extent of simultaneous prescribing of atypical antipsychotics, conventional antipsychotics and anticholinergics in daily clinical practice

Pharmacoepidemiological study of pharmacists in 45 hospitals in Belgium, Denmark, France, Germany, the Netherlands, and Scotland

11/1998 – 10/1999

Prescription data

2,725 patients using AAPs for more than six weeks

Frequencies of simultaneous prescription of atypical anti-psychotics with other anti-psychotics and/or anticholinergics

Gives range of frequencies of atypical polypharmacy only: between 26.1% (Germany) and 49.1% (Belgium); without lowpotent APs they vary between 17.1% (Denmark) and 31.6% (Belgium)

Study not designed to detect differences in AAP use in different countries (some countries underrepresented)

[7]

(1) To analyse the trend of antipsychotic drug consumption in Spain, comparing it with that of Nordic countries (2) To evaluate the impact of the introduction of newer AAPs To compare the prevalence of psychotropic medication use in youth

Pharmacoepidemiological study of national drug consumption

1985– 2000

Retail community pharmacy sales as recorded in Spanish Ministry of Health database

Populations of Spain and of five Nordic countries as comparator

DDD/1,000/d

2000: 5.73 (all antipsychotic drugs) 1.31 (risperidone) 1.21 (olanzapine) ~0.1 (clozapine) 0 (sertindole, quetiapine)

Data on trends do not include hospital use and so might underestimate trends in consumption over time and across countries

[8]

Cross-sectional population-based analysis of administrative claims data in three countries

2000

Administrative claims data

Insured outpatient youth aged 0–19 in the Netherlands, Germany, and the USA

Annual prevalence of use defined as the dispensing of one or more prescriptions for a psychotropic drug per 100 enrolled youth

% AAP to total antipsychotic use was 5% in Germany, 48% in the Netherlands, and 66% in the USA

[9]

To evaluate facets of psychiatric inpatient care of patients in a German and a Japanese hospital

Pharmacoepidemiological study of patients with schizophrenia admitted to two hospitals each in a different country

1997

Psychiatric basic documentation system

865 German inpatients and 50 Japanese inpatients with schizophrenia

Frequencies of psychopharmacological treatment

The rate of AAP use was 18% in Germany (clozapine) and 12% in Japan (zotepine)

US data based on one state only so generalisability of findings may be limited; lack of diagnostic data does not allow inference about indication of prescribed medicines Unit of analysis: single hospitals in two countries limiting the generalisability of findings

(Data on AAP use in Nordic countries not reported )

National sales data for Nordic countries

8

[10]

International variation in the usage of medicines

Atypical anti-psychotics

Table 1 Summary of studies of international variation in the use of atypical anti-psychotics (AAPs) - continued Stated study objective

Study design

Year

Data source

Population/s studied

Outcome measure/s

Key findings

Notes

To compare drug prescriptions between German and Swiss psychiatric services with regard to their preference of newer psychotropics

Pharmacoepidemiological study of psychiatric inpatients in five hospitals in Germany (n=3) and Switzerland (n=2)

1999

Administrative data on drug prescriptions

572 psychiatric inpatients

Proportion of patients receiving psychotropic drugs including AAPs (clozapine, olanzapine, risperidone, and amisulpride)

Proportion of patients receiving AAP varied between 25% and 45% but this variation was not significant and did not differ by country; proportion of patients receiving clozapine was higher in German hospitals (at 70.8% of prescribed AAP compared with 43.9% in Swiss hospitals)

Data were adjusted by age and sex; although the five hospitals were selected to represent university and non-university settings as well as urban and rural areas, generalisability of findings is limited

[11]

To describe the baseline findings and study population of the Schizophrenia Outpatient Health Outcomes (SOHO) Study

Three-year prospective, observational study of the health outcomes associated with antipsychotic treatment in ten European countries (Denmark, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, and the UK)

Recruit ment: 9/2000– 12/2001

Core data collection form (DCF); data collected by participating psychiatrists (n=1096)

10,972 outpatients with schizophrenia who initiate therapy or change to a new anti-psychotic

Anti-psychotic medication along with a wide range of demographic, clinical, and social indicators

Atypical anti-psychotic treatment in the six months prior to inclusion in the study varied between countries: Denmark (60.6%), Portugal (58.1%), Ireland (43.6%), the UK (42.9%), Spain (40.7%), France (401.%), the Netherlands (39.2%), Italy (38.5%), Germany (34.2%), and Greece (27.4%); average: 37.6%

Analyses baseline characteristics of population recruited into the study; study not designed to analyse variation in AAP use

[12]

To compare prescription practices for acute inpatients with schizophrenia among six academic departments in four countries

Pharmacoepidemiological study of psychiatric inpatients in six academic departments in China (Hong Kong; Shanghai), Japan (Tokyo), Hungary (Budapest; n=2), and the USA (New York)

1999

Standardised data collection form

429 inpatients with clinical diagnosis of schizophrenia

Number and dose of AP, AAP and depot APs; other psychotropic drugs; and multiple APs

Proportion of patients receiving AAP was highest in one Budapest centre: 73.1% and Shanghai: 74.6%; Hong Kong: 38%; Budapest 2: 36%; New York: 42.7%; Tokyo: 8.9%

Unit of analysis: single hospitals in four countries limiting the generalisability of findings

[13]

9

Source

CHAPTER 4

Dementia

The PubMed search yielded 217 records, of which two met the inclusion criteria. In addition we searched the websites of the following organisations: European Dementia Consensus Network, Alzheimer’s Research Trust, Alzheimer Europe, the National Institute on Ageing, For Dementia, and the European Federation of Pharmaceutical Industries and Associations (EFPIA), but this search did not identify additional studies considered eligible. Pariente et al (2008) examined the prevalence of cholinesterase inhibitor (ChI) treatment in nine European countries (Belgium, France, Germany, Italy, the Netherlands, Poland, Portugal, Spain and the United Kingdom) in 2004 (Table 2).14 Use of ChIs (donepezil, rivastigmine, and galantamine) in dementia patients varied greatly across countries, ranging from a low in the Netherlands (3.0 per cent), to 5.9 per cent in Italy, 6.7 per cent in the United Kingdom, 17.5 per cent in Spain and 20.3 per cent France. For those treated with a single drug, donepezil was the most frequently used ChI across all countries except the Netherlands, where this drug was not marketed at the time of the study (2004); use ranged from 47 per cent in Spain to 72 per cent in the United Kingdom. The proportion of those using rivastigmine was relatively higher in France (20 per cent), Belgium (26.4 per cent), Germany (22.5 per cent), and the United Kingdom (17.3 per cent), but very low in Poland (0.2 per cent). Galantamine was used more frequently in Spain (33.5 per cent), Portugal (24 per cent), Poland (30.4 per cent), Italy (25.7 per cent), and the Netherlands (85 per cent), but less so in the United Kingdom (10.6 per cent), Germany (14 per cent), and France (16.2 per cent). The authors discussed a range of potential factors that might explain variation in dementia drug use, in particular the role of what they broadly term “health policies”, in this instance referring to reimbursement policies. For example, countries such as the Netherlands require a complex negation process for reimbursement involving clinical, biological, and radiological information for each patient who required dementia medication. Reimbursement of national health insurance systems varied and ranged from 0 per cent in Italy to 100 per cent in the United Kingdom. As the initial prescription of ChIs is restricted to specialists such as neurologists, psychiatrists, and geriatrists, variation in usage might also be caused by availability of specialists.

11

International variation in the usage of medicines

International Healthcare Comparisons

Table 2 Summary of studies of international variation in the use of medicines for the treatment of dementia Stated study objective

Study design

Year

Data source

Population/s studied

Outcome measure/s

Key findings

Notes

To evaluate the prevalence of cholinesterase inhibitor (ChI) treatment in subjects with dementia in European countries

Pharmacoepidemi ological study of prevalence of ChI use

2004

Sales (IMS Health) and reimbursement data (donepezil, rivastigmine, and galantamine); estimates of dementia prevalence

Subjects with dementia in nine European countries (Belgium, France, Germany, Italy, the Netherlands, Poland, Portugal, Spain, and UK)

Prevalence of treatment in subjects with dementia

Prevalence of treatment with ChIs varied: Netherlands (3.0%), Italy (5.9%), the UK and Germany (6.7%), Poland (7.9%), Portugal (11.2%), Belgium (13.8%), Spain (17.5%), and France (20.3%)

Potential limitations of findings: (1) rely on ecological data for the prevalence of dementia and drug utilisation; (2) noted inconsistencies of IMS data in relation to dosage estimates

[14]

To examine access to ChIs across the European Union

Cross-sectional survey of clinical experts in 23 countries

2005

Survey of accessibility of four dementia drugs (rivastigmine, galantamine, donepezil, and memantine)

One clinical expert in each of 23 countries(Austria, Belgium, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Lithuania, Luxemburg, Malta, the Netherlands, Poland, Portugal, Slovakia, Slovenia, Spain, Sweden, Switzerland, and the UK)

(1) drug is licensed (yes/no) (2) drug is reimbursed (fully/partial/no) (3) GPs are allowed to initiate (yes/no) and continue treatment (yes/no)

Variability in accessibility of drug treatment across EU. Only six countries (Switzerland, Ireland, Sweden, Malta, Germany, and Luxemburg) allowed GPs to initiate dementia treatments. Variability in reimbursement: 12 of the 23 countries provided full reimbursement for all four of the drugs studied; variation in restrictions posed on reimbursement from none (e.g. Ireland and Switzerland) to various (the Netherlands)

Survey includes one expert per country who had to meet certain requirements (clinical duties in patient care for people with dementia; authored at least one publication regarding dementia in a peerreviewed journal in 2004 or 2005); representativeness of findings might be questionable

[15]

12

Source

International variation in the usage of medicines

Dementia

For example, the authors argued that the “relative lack” of relevant specialists in the United Kingdom could partly explain the low rates of ChI use in the country, despite the full reimbursement of treatment costs for patients. However, Pariente et al. (2008) stressed that although health policies constitute an important determinant of observed differences, they do not explain all variations in ChI usage. Other factors to be taken into account include variation in diagnosis of dementia, and the authors pointed to the introduction of a plan of action for Alzheimer’s disease in France in 2001, which aimed at improving dementia detection and treatment and which is likely to have increased the number of people diagnosed with the disease and, consequently, the number of patients receiving treatment, although data to confirm this hypothesis could not be presented. Other potential reasons include differences in prescribing patterns among European countries (although not specific to dementia drugs), citing evidence that the more than 70 per cent of medical consultations tend to result in a prescription in countries such as France and Germany while occurring in less than half of consultations in the Netherlands, for example.14 Oude Voshaar et al. (2006) examined the accessibility of donepezil, rivastigmine, memantine, and galantamine for patients with dementia across Europe by surveying clinical experts in 23 European countries in 2005 (Table 2).15 “Clinical expert” was defined as a clinician who (1) had to have clinical duties in patient care for people with dementia and (2) had authored at least one publication on dementia in a peer-reviewed journal in 2004 or 2005. Other than galantamine in Hungary and donepezil in the Netherlands, all of the medicines under review had been licensed throughout the EU. However, countries varied with regard to the indications for which dementia medicines were licensed. For example, memantine was licensed for mild to moderate Alzheimer’s disease in Slovenia, Slovakia, and Malta only, but it was licensed for moderate to severe Alzheimer’s disease in all the other countries included in the survey. Oude Voshaar et al. (2006) demonstrated that countries varied greatly with regard to the role of the GP in initiating dementia treatment, which was only authorised in six countries (Switzerland, Ireland, Sweden, Malta, Germany, and Luxemburg), while eight countries prohibited GPs from continuing anti-dementia drug treatment initiated by specialists (Netherlands, Belgium, France, Slovakia, Czech Republic, Austria, Hungary, and Portugal).15 Reimbursement rates also varied across countries. Only 12 of the 23 countries studied provided full reimbursement for all four of the medications studied, with the remainder providing only partial reimbursement. Galantamine was not reimbursed in four countries (Hungary, Lithuania, Malta, and Poland), memantine in three countries (Hungary, Malta, and Poland), rivastigmine in two countries (Lithuania and Malta), and donepezil in two countries (Malta and the Netherlands). In addition, countries differed in the restrictions they imposed on reimbursement, with some imposing no restrictions (e.g. Ireland and Switzerland), while others based reimbursement decisions on treatment protocols for Alzheimer’s disease (e.g. the Netherlands and Belgium). Several countries also required permission from the relevant funder prior to treatment (Netherlands, France, Luxembourg, and Denmark). In addition, in the Netherlands, although memantine was licensed for moderate to severe dementia, it was only reimbursed for severe dementia.

13

International variation in the usage of medicines

International Healthcare Comparisons

This study provides important insights into the possible causes of variation in dementia drug use across European studies. For example, the findings for the Netherlands in relation to restrictions imposed on reimbursement might explain, in part, the low use of ChIs in that country; France also required prior approval yet the use of dementia medicines was relatively high (Table 2). It should however be emphasised that the survey of clinical experts was undertaken in 2005 and it is likely that countries’ policies on treatment pathways and accessibility of drugs under the statutory system has changed since.

14

CHAPTER 5

Rheumatoid arthritis

The PubMed search identified 955 titles of which four studies met the inclusion criteria. The websites of the following organisations were searched for grey literature: Arthritis Care, the National Rheumatoid Arthritis Society, Arthritis Research Campaign, Rheumatology Information Service Europe, and the European Federation of Pharmaceutical Industries and Associations (EFPIA), the latter identifying one additional document.16 Of the four studies identified, Sokka et al. (2007) presented the findings of a cross-sectional review of the treatment of unselected consecutive outpatients with rheumatoid arthritis in 15 countries.17 However, although this study analysed the use of disease-modifying antirheumatic drugs (DMARDs) it did not disaggregate for the new biological drugs (TNF inhibitors) of interest here and we therefore did not include this study in our review. Jönsson et al. (2008) examined international variation in use of TNF inhibitors (etanercept, infliximab, and adalimumab) and of conventional DMARDs (aurothiomalic acid, auranofin, and leflunomide) (Table 3).18 The study examined the period 2000 to 2006 in 30 countries and showed that the United States had the fastest and most extensive uptake of TNF drugs as measured by sales data. By the end of 2006, the US rate was in excess of €350,000 per 100,000 population (data are presented as graphs only with exact figures not reported), approximately three times the average in western European countries and Canada. High uptake was also observed for Norway (~€350,000/100,000), Sweden (~€300,000/100,000), the Netherlands (~€250,000/100,000), and Finland (~€200,000/100,000). France, Spain, and the United Kingdom were approximately at the E-13 average (defined as EU-15 countries excluding Portugal, Ireland, Greece, and Luxemburg, but with the addition of Norway and Switzerland), at around €130,000 per 100,000 population, while Germany (~€110,000/100,000) and particularly Italy (~€70,000/100,000) were below average, as was Australia (~€20,000/100,000). Indeed, the level of uptake seen in Australia was more akin to the levels observed for the ten countries of central and eastern Europe that were also included in the study, at an average level of €10,000/100,000. Within the (western) European market, and after correcting for differences in the prevalence between northern and southern Europe, by the end of 2006, France and Spain had the highest uptake, at around €550 per patient with rheumatoid arthritis (RA); uptake was lower in the United Kingdom (~€400/patient with RA), Germany (~€350/patient), and Italy (~€300/patient) (E-13 average: ~€400/patient). The estimated number of patients per 100,000 population treated with TNF inhibitors was highest in the United States (around 140 per 100,000 population) and Norway

15

International variation in the usage of medicines

International Healthcare Comparisons

(~120/100,000). The lowest rates were seen in Australia, Slovenia, and the Czech Republic at around or under 10/100,000. When disaggregated by individual TNF inhibitor, the relative “ranking” of countries changed somewhat, with for example the United Kingdom well above the E-13 average for the use of etanercept (~€120/patient with RA compared with E-13 average at €90/patient) but lower for infliximab (~€45/patient compared with €60/patient) and, in particular, adalimumab (~€30/patient compared with €60/patient). In contrast, Germany was shown to be well below the E-13 average (and the United Kingdom) for the use of etanercept (~€80/patient) and infliximab (~30/patient) but well above for the use of adalimumab (~70/patient). In trying to explain observed variations in the uptake of TNF inhibitors, Jönsson et al. (2008) noted differences in national income (as measured by GDP) as a possible reason for the much lower use of these medicines in the countries of central and eastern Europe compared with western Europe, Canada, and the United States.18 However, at the same time there were also considerable differences between countries with broadly the same GDP. Relative price levels might explain some of the variation, with, for example, the relatively high price in Germany believed to have affected access. Yet, although there were substantial differences in prices for TNF inhibitors, high and low use was not systematically related to differences in price. The authors highlighted the possible role of health technology assessment (HTA) in uptake, noting that countries with a strong HTA tradition, combined with a societal perspective, tended to show greater uptake of TNF inhibitors such as Sweden and Norway compared with countries that tend to apply stricter health care cost approaches such as the United Kingdom and Australia, and so concluding that international variation in the uptake of these medicines reflects national preferences and priorities.18 However, the authors emphasised that the nature of the data examined did not allow for establishing a clear cause–effect relationship of this hypothesised association. They further pointed to a potential role of variations in access to specialists (rheumatologists), with an estimated density of 1/25,000 population in France and 1/50,000 in the United States, compared with only 1/150,000 in the United Kingdom and 1/200,000 in Germany, which might explain some of the observed variations in uptake between countries.

16

International variation in the usage of medicines

Rheumatoid arthritis

Table 3 Summary of studies of international variation in the use of medicines for the treatment of rheumatoid arthritis (RA) Stated study objective

Study design

Year

Data source

Population/s studied

Outcome measure/s

Key findings

Notes

To compare uptake of RA treatment across countries

Pharmacoepidemiological study of uptake of TNF inhibitor in 30 countries

2000– 2006

Commercial databases (IMS Health) (etamercept, infliximab, and adalimumab)

Australia, Austria, Belgium, Bulgaria, Canada, Croatia, Czech Republic, Denmark, Estonia, Finland, France, Germany, Ireland, Italy, Latvia, Lithuania, the Netherlands, Norway, Poland, Romania, Russia, Slovak Republic, Slovenia, South Africa, Spain, Sweden, Switzerland, Turkey, the UK, and the USA

Total sales (in €) per 100,000 population for class and individual TNF inhibitor; “crude” rate and adjusted for disease prevalence; estimated number of patients receiving TNF inhibitor treatment

Estimated number of patients per 100,000 population treated with TNF inhibitor was highest in USA (~140 per 100,000 population) and Norway (~120/100,000); followed by Sweden and Netherlands (~75–80); Denmark, Belgium, Finland, and Luxembourg (~60); Canada (~50); UK and Switzerland (~40); Spain, France, and Italy (~35); Germany (~25); Austria (70%: Hungary, Latvia, the Netherlands, and Turkey: 65%; Bulgaria: 59%

Survey population identified from selected geographical areas and mainly academic hospitals so not representative of all CHD patients in given country

Source

[27]

[28]

To compare the Irish results of EUROASPIRE III with the rest of Europe

Cross-sectional survey of patients with coronary heart disease (CHD) in 22 countries

2006– 2007

Patient medical records and patient interview and examination

386 Irish patients with first or recurrent diagnosis or treatment of CHD in two Dublin centres; nonIrish patient population of EUROASPIRE III (figure not given)

26

Reported medication (cardioprotective drugs), prevalence of CHD disease risk factors; therapeutic control of blood pressure and blood sugar

Use of statins was significantly higher in Ireland than in the rest of the EUROASPIRE III population, at 91% compared with 79%

Survey population identified from two centres in Dublin and so not representative of all CHD patients in Ireland

[29]

CHAPTER 7

Hepatitis C

The PubMed search identified 91 records of which one study met our inclusion criteria. We also searched the website of the European Federation of Pharmaceutical Industries and Associations (EFPIA). Lettmeier et al. (2008) examined the market uptake of peginterferons for the treatment of hepatitis C in 21 European countries during 2000–2005.30 The study found that peginterferon sales (per 1,000 population) varied by region, with highest sales seen among the group of EU Founding Members (Belgium, France, Germany, Italy, and the Netherlands), which were also found to be the earliest and most rapid adopters of peginterferons. These were followed by the group of countries that had joined the EU before 2000 (Austria, Denmark, Finland, Greece, Ireland, Spain, Sweden, and the United Kingdom) and those joining after 2000 (Czech Republic, Hungary, Poland, and Romania). The lowest sales rates were seen among the combined group of four non-EU countries (Norway, Russia, Switzerland, and Turkey). By 2005, an estimated total of 308,000 patients had received treatment with peginterferons in the 21 countries included in the study. The number of those ever treated varied widely, from a high of 16 per 100 prevalent cases in France, followed by the Czech Republic, Germany, the Netherlands, and Sweden, at between 10 and 12 per cent, to less than 1 per cent of cases in Greece, Poland, Romania, and Russia. The United Kingdom was among countries with a relatively low number of patients treated, at around 3.5 per cent, equating to the average rate in the 21 countries. In an attempt to explain observed variations, the authors point to the potential role of financial restrictions, with sales rates found to be low in countries imposing restrictions on reimbursement of treatment costs, such as in some countries of central and eastern Europe. When considering the variation in the number of patients ever treated, Lettmeier et al. (2008) highlighted the role of under-detection of prevalent cases, citing evidence that in France, which operates an active screening policy for Hepatitis C, about 40 per cent of cases remain undetected whereas in Spain this figure is estimated at 80 per cent.30 The authors further point to the potential impact of drug policies that might restrict access to treatment for high risk groups such as injecting drug users, although these considerations are not systematically followed up.

27

International variation in the usage of medicines

International Healthcare Comparisons

Table 6 Summary of studies of international variation in the use of new anti-viral drugs for the treatment of hepatitis C Stated study objective

Study design

Year

Data source

Population/s studied

Outcome measure/s

Key findings

Notes

To compare the market uptake of peginterferons across countries of the WHO European region

Pharmacoepidemiological study of the uptake of peginterferon alpha-2a and alpha-2b

2000– 2005

IMS Health database (quarterly sales data)

EU Founding Members: Belgium, France, Germany, Italy, and the Netherlands EU-joiners before 2000: Austria, Denmark, Finland, Greece, Ireland, Spain, Sweden, and the UK EU-joiners after 2000: Czech Republic, Hungary, Poland, and Romania Non-EU: Norway, Russia, Switzerland, and Turkey

Country-specific sales rates; annual number of patients treated (per population and per prevalent hepatitis C case)

Peginterferon sales varied by region, with highest sales among EU Founding Members and lowest among non-EU countries; countryspecific cumulative treatment rates per 100 prevalent cases were 16 (France); 12 (Germany and Sweden); ~11.5 (Czech Rep and Netherlands);